Page 24 - Handbook of Thermal Analysis of Construction Materials

8 Chapter 1 - Thermoanalytical Techniques

cases, the amplitude of the signal of dissipated power is affected by the
heating and cooling rates. Based on these facts, it is obvious that the
accuracy of the measurement is generally lower than the degree of
reproducibility.
There are quite a few different methods for the calibration of DSC
instruments, of which the most popular are: (a) calibration by Joule-effect
and (b) calibration by heats of fusion. [12][15][30] The Joule-effect calibration
is relatively simple and straight-forward in that it consists of an electrical
heater inserted into the sample and reference compartments. A pulse of
predetermined duration and intensity is sent to the sample, and the dissi-
pated power is then measured. The disadvantage of this method is that some
heat flux can be dissipated in the heater wires, and, therefore, not truly
measured. Furthermore, the electrical heater is not necessarily composed of
the same material as the sample and reference holders. Still, the accuracy
of this calibration technique is better that 0.2%.
The heats of fusion calibration method affords two simultaneous
calibrations. Pure substances, which undergo phase transformations at very
well-characterized temperatures, are used. Since the enthalpy of fusion and
temperature of fusion of the calibrant are well known, both a temperature
and enthalpic calibration can be performed with the same substance.
Ideally, more than one compound and more than one scanning rate should
be utilized (or if only one scanning rate is employed, then the scanning rate
should correspond to that which will be used for the experiment) since the
sensitivity of the measurement is not only temperature dependent, but also
scan rate dependent. Since the thermal conductivity might play an impor-
tant role in the measured response, the mass of the calibrant should be as
close as possible to the sample mass. The following criteria should be used
when choosing a calibrant:
a) The substance must be available in high purity.

b) The transition temperature and enthalpy of transition
should be known with a high degree of accuracy.

c) The substance should not show any tendency to super-
heating. [4][5][12]

The major drawback of this method is that since transitions are very
temperature specific, one substance might be suitable for only one tempera-
ture range, hence the need to use more than one calibrant (or one must
assume that the calibration will hold for the entire range being studied).
Another calibration method is with the use of radioactive materials